sign in sign up
<<
Home , Adar, ADARB2

r e v i e w

Deciphering the functions and regulation of brain-enriched A-to-I RNA editing

Jin Billy Li1 & George M Church2

Adenosine-to- (A-to-I) RNA editing, in which genomically encoded is changed to inosine in RNA, is catalyzed by acting on RNA (ADAR). This fine-tuning mechanism is critical during normal development and diseases, particularly in relation to brain functions. A-to-I RNA editing has also been hypothesized to be a driving force in human brain evolution. A large number of RNA editing sites have recently been identified, mostly as a result of the development of deep sequencing and bioinformatic analyses. Deciphering the functional consequences of RNA editing events is challenging, but emerging genome engineering approaches may expedite new discoveries. To understand how RNA editing is dynamically regulated, it is imperative to construct a spatiotemporal atlas at the species, tissue and levels. Future studies will need to identify the cis and trans regulatory factors that drive the selectivity and frequency of RNA editing. We anticipate that recent technological advancements will aid researchers in acquiring a much deeper understanding of the functions and regulation of RNA editing.

One striking observation from the Project is that ADAR1 and ADAR2, but not ADAR3 (also known as ADAR, ADARB1 only ~20,000 -coding are present in humans, a surpris- and ADARB2, respectively)5. In mice, ADAR1 is required for embry- ingly low number that does not scale with human developmental and onic development, as Adar1−/− embryos die between E11.5 and E12.5 cognitive complexity1. Part of the answer to this apparent paradox lies as a result of failed hematopoiesis maintenance6,7. ADAR2 is also in the complexity of RNA, including alternative splicing and non- required, as null mutants develops epileptic seizures and die several coding RNA. RNA editing, a co-transcriptional process in which weeks after birth8. Although ADAR1 and ADAR2 are expressed the genome-encoded information is altered in RNA, provides a in many tissues, the inactive ADAR3 is expressed exclusively in potentially powerful method for diversifying the transcriptome and the brain5. It is intriguing to speculate on potential functions of fine-tuning biological function2. ADAR3 in the brain. © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature A-to-I editing, the most common type of editing known in , leads to the recognition of inosine as by the translation, Dysregulation of RNA editing in neurological diseases splicing and sequencing machineries2. A-to-I editing is carried out by Aberrant RNA editing has been linked with a variety of diseases, pri- npg the ADAR family of enzymes2, which are conserved across metazoans. marily neurological or psychiatric disorders. As of several years ago, All ADARs share double-stranded RNA (dsRNA)-binding domains only a handful of RNA editing sites had been identified and examined; and catalytic deaminase domains that deaminate A to I. Here we dis- however, these sites were found to be dysregulated in many diseases2,9. cuss recent progress and future needs for studying A-to-I RNA editing For example, the editing level (fraction of edited transcripts) of the from the perspectives of identification, evolution, function, atlas and glutamate receptor GluA2 (also known as GluR2, GluR-B and Gria2) regulation (Fig. 1). Q/R site might be altered in epilepsy, amyotrophic lateral sclerosis, malignant glioma, schizophrenia and ischemia2,9. In the serotonin ADAR mutants exhibit neural and behavioral phenotypes receptor 5-HT2CR, there are five editing sites located within 32 bases, 10 Inactivation of ADAR family members results in mainly neuro- giving rise to 24 protein isoforms . Dysregulated 5-HT2CR (HTR2C) nal and behavioral phenotypes. Knockout of the single Drosophila mRNA editing might be involved in depression and suicide, schizo- melanogaster ADAR leads to brain-related phenotypes such phrenia, Prader-Willi syndrome, and metabolic diseases such as obes- 2,9,11 as uncoordinated locomotion, temperature-sensitive paralysis ity and diabetes . Although RNA editing in GluA2 and 5-HT2CR and age-dependent neurodegeneration3. When both of its ADAR has functional consequences under physiological conditions, the man- genes are deleted, Caenorhabditis elegans exhibits defective ner in which altered editing might be involved in disease pathogenesis chemotaxis4. Mammalian genomes carry three members of the needs to be further studied. More recently, several studies examined a ADAR family, and enzymatic activity has been demonstrated for larger number of sites, and the editing level of a number of these sites were found to be altered in schizophrenia, bipolar disease12, autism13 14 1Department of Genetics, Stanford University, Stanford, California, USA. and cancers , although replication experiments with a larger cohort 2Department of Genetics, Harvard Medical School, Boston, Massachusetts, USA. may be needed to confirm these findings. Correspondence should be addressed to J.B.L. ([email protected]). Although these results establish links between RNA editing Received 11 July; accepted 11 September; published online 28 October 2013; and a number of diseases involved with the nervous system, build- doi:10.1038/nn.3539 ing a functional relationship requires the following considerations.

1518 VOLUME 16 | NUMBER 11 | NOVEMBER 2013 nature neuroscience r e v i e w

a large number of the sites identified were noncanonical RNA editing Mining RNA-seq data events, which were shown to be mainly technical artifacts23–25. In our work, we were unable to identify and confirm these noncanoni- Sites cal RNA editing sites26,27. The false discovery of noncanonical sites is mainly derived from errors in the sequencing library preparation and the sequencing process, as well as the computational mapping of Atlas Evolution RNA sequencing (RNA-Seq) reads to the genome and transcriptome. Spatiotemporal map at In comparison with mapping genome sequencing reads, it is more Increased neural function RNA species, tissue, and cell levels and human cognition? editing challenging to map RNA-Seq reads to identify variants at the single- base resolution for two main reasons: the errors introduced by the random hexamer that is often used to make RNA-Seq libraries and the various RNA splicing events expressed at different levels23–25. All Function Regulation of the methods require both matched genome sequencing data along

cis: motif + RNA structure; with the RNA-Seq data to effectively distinguish RNA editing sites Enabled by new technologies trans: ADARs + other factors from SNPs. However, many of the samples with RNA-Seq data did not have the genome sequencing data. In fact, samples with matched Figure 1 Overview. We highlight the major, but not all, discussion topics. genome and RNA sequencing data were mostly lymphoblastoid cell lines that were used in the 1000 Genomes Project. To fully utilize publicly available RNA-Seq data, such as those First, with the recent explosion of newly identified RNA editing from brain tissues where RNA editing is more prevalent, we recently sites (see below), a comprehensive screening of many, if not all, of developed an approach for identifying RNA editing sites using only the sites in disease contexts will provide an unbiased assessment. RNA-Seq data, without matched genome sequencing data. To dis- Second, a large number of cases and controls are needed to achieve tinguish RNA editing sites from SNPs, we developed two different high statistical power, particularly to pinpoint relatively subtle dif- approaches28. In the first approach, multiple human RNA-Seq data ferences. Third, models of disease may be helpful because sets are used, RNA variants are called and known SNPs are removed. fair comparisons are possible between ‘cases’ and controls that share RNA editing events are enriched because rare SNPs are unlikely to be the same genetic background and environmental conditions. Fourth, shared by different individuals. In the second approach, RNA-Seq data functional in vitro or in vivo assays are needed to perturb the edit- sets from related species are used. The RNA editing sites conserved ing levels and correlate with disease phenotypes, as exemplified in between species are identified because SNPs independently occur previous work8,11,14. in different species and are therefore not shared between species. Using these strategies allowed us to substantially expand the number Identification of RNA editing sites: a rapid expansion of identified editing sites, as our methods permitted us to analyze a The first handful of mammalian A-to-I RNA editing sites were iden- large number of samples for which RNA-Seq data, but not genome tified serendipitously in glutamate and serotonin receptors10,15. The sequencing data, exists. difficulty in finding other editing sites was primarily a result of the When RNA editing sites are identified using computational limitations of the sequencing technology at the time, particularly in approaches described above, one often requires a high fraction (for © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature generating the coverage necessary to delineate the single-base dif- example, >80%) among all variants to be A-to-G variants, indicative ferences between genomic DNA and RNA. In the early 2000s, large of A-to-I RNA editing. At the 80% A-to-G fraction, the false discov- sequencing efforts to generate expressed sequence tags (ESTs) of ery rate is estimated to be ~2% (refs. 26,28). To further confirm the npg genomes and transcriptomes made it possible to identify RNA editing computational predictions, biochemical evidence will be needed. For sites across the transcriptome. This is exemplified by searches using example, using RNA-Seq data from Drosophila wild-type and ADAR comparative genomics approaches in Drosophila16 and mammals17. knockout lines, we estimated that the false discovery rate of our However, the success of these screens was limited, mostly as a result computational prediction was 1.8% (ref. 28). of the relatively small amount of data available at the time and the Over 1.4 million human RNA editing sites have been identified difficulty of distinguishing RNA editing sites from genomic single (this number is expected to continue growing) and the vast majority nucleotide polymorphisms (SNPs). In contrast with this limited suc- of them are located in noncoding sequences in Alu repeats28. Only cess in identifying nonrepetitive sites, several groups identified thou- about 200 sites have been identified in nonrepetitive protein cod- sands of clustered editing sites in Alu repeats18–20. Primate-specific ing regions, of which ~60% result in amino acid changes. The most Alu repeats are heavily edited because they comprise ~1.2 million recently identified recoding sites tend to have lower editing levels, copies occupying >10% of the human genome, leading to prevalent suggesting that recoding sites that are highly or moderately edited double-stranded RNA when in inverted repeats. However, the func- are saturated. However, these lowly edited sites could be highly edited tion of Alu editing sites is largely unknown. in specific tissues and/or cell types. Sites that are variably edited in Next-generation sequencing is particularly helpful for compre- different tissues or cell types may have specialized functionality and hensively identifying RNA editing sites. In the early days of next- therefore be potentially interesting. generation sequencing, when whole human genome and transcriptome The recoding RNA editing sites are enriched in neuronal genes. sequencing was not feasible, we developed a targeted capture tech- This is not observed for the set of all genes containing RNA editing nique to sequence ~36,000 human sites that were computationally pre- sites in Alu repeats. However, if one focuses on the subset of human dicted to be candidates of RNA editing sites and identified hundreds of genes harboring Alu repeat editing sites that are conserved between new A-to-I sites21. Recently, several groups independently developed human and other primates, they are also enriched for neuronal func- methods for identifying RNA editing sites by comparing the genome tions28. Future work is needed to determine how the editing events and RNA sequencing data of the same individuals22. Unfortunately, in neuronal genes may affect neural functions.

nature neuroscience VOLUME 16 | NUMBER 11 | NOVEMBER 2013 1519 R e v i e w

Number of editing sites conserved in human splicing, trafficking, microRNA binding and ribosome binding effi- 0 5,000 10,000 15,000 20,000 Human ciency. However, very little is currently known in the field, probably 20 Myr Chimpanzee as a result of the technical challenges of deciphering the functions of an individual RNA editing event. Most of the methods widely used to Rhesus macaque study gene function (for example, RNA interference) are not suited to Mouse studying the function of an editing site. Overexpression of edited and unedited isoforms in a heterologous system provides an important first step in regards to examining the functional consequences of RNA 0 100 200 300 400 500 600 700 editing events30. Until now, the gold-standard method of choice was Alu sites Repetitive non-Alu sites to generate a knock-in animal with either abolished or constitutive Nonrepetitive sites (non-coding) RNA editing4,11. Because RNA editing occurs in dsRNA structures Nonrepetitive sites (coding) recognized by ADARs, the flanking sequence of an editing site has to Figure 2 Decreased number of RNA editing sites shared with human sites be paired with an editing complementary sequence. To abolish RNA with increased phylogenetic distance. Phylogenetic relationships between editing in a coding exon, one can remove the editing complemen- human, chimpanzee, rhesus macaque and mouse are shown on the left. tary sequence that may be present in the intron, thereby not directly Myr, million years ago. Numbers of A-to-I editing sites that are conserved affecting the coding regions. To produce constitutive RNA editing, between human and each of the other three species are shown on the right, one would change the editing site from adenosine to guanosine at the with the numbers in non-Alu regions magnified at the bottom. The number of editing sites conserved in human (indicated in the scales) is based on genomic DNA level. One classic example is the Q/R site in the GluA2 2+ the analysis of the same number of reads from each species28. Figure is gene; when editing is abolished in mice, excess influx of Ca into adapted from ref. 28. neurons leads to seizures and postnatal death31. Another example is the G protein–coupled serotonin receptor 5-HT2CR, which has five recoding sites located within 13 bases. In knock-in mice with consti- RNA editing as a driving force of brain evolution? tutive RNA editing, the sympathetic nervous system is constitutively Although many RNA editing sites, particularly recoding sites, activated, the energy expenditure is enhanced and the receptor–G are shared between human and mouse, the intensity of RNA edit- protein coupling is blunted, which is masked by a higher density of ing in human is 35-fold higher than that in mouse19. This is a result receptor binding sites11,32. Despite previous success, the conventional of the widespread RNA editing derived from primate-specific Alu approach is very tedious, time-consuming and costly. Recent techno- sequences. In primates, the amount of editing appears to have logical advancements in genome engineering, such as TALEN and increased during primate evolution, based on the examination of six CRISPR-based approaches33, have the potential to expedite studies Alu repeats29. Human and non-human primates have highly similar using the knock-in mouse approach. genomes, but differ substantially in cognitive capacities. Given the Given the fact that most human RNA editing sites are not present findings of this recent study and the important roles of RNA editing in mouse28, it is important to establish in vitro methods using in brain-related functions, it is plausible that the expansion of RNA human cells as the model system. In addition to using the TALEN editing in humans may have led to increased cognition and driven and CRISPR-based approaches, a recently developed method using neural evolution. an antisense oligoribonucleotide was demonstrated to effectively An unbiased, large-scale comparison of RNA editing events between inhibit RNA editing34. Alternatively, overexpression of either edited © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature human and non-human primates is needed to test this hypothesis. In or unedited transcripts may also lead to informative phenotypes14. a comparative transcriptome study, we found that the number of RNA editing sites in a given species that are shared with human editing sites Toward a spatiotemporal atlas of RNA editing npg decreases with increased phylogenetic distance28 (Fig. 2). Although our RNA editing is known to be a dynamically regulated process21,35,36. data indicate that many human RNA editing sites are absent in non- Constructing a spatiotemporal quantitative atlas of RNA editing human primates, they cannot exclude the possibility that many of the would be a first step in understanding its dynamic regulation. The non-human primate sites are also absent in human. Studying the evo- recent explosion of identified RNA editing sites in human and other lution of RNA editing will reveal not only the functional sites that are species empowers a large-scale study to create an atlas of RNA edit- conserved between species, but also the sites gained, lost or differentially ing. RNA sequencing allows for measurement of editing frequency at edited across species. This provides a list of plausible candidates for future the transcriptome level, although the accuracy may be compromised study for their potential roles in human brain cognition. Furthermore, for sites in lowly expressed genes. Alternatively, targeted approaches this rich data set, when generated, will facilitate understanding of why to sequences only those regions that harbor RNA editing sites (par- editing events differ between human and other primates. ticularly those in functionally important regions) would enhance the In addition to studying natural variations in RNA editing between sequencing accuracy in lowly expressed genes and lower the cost21,35, humans and other primates, a complementary approach is to alter although current techniques need to be improved. RNA editing using genome engineering approaches. For example, Previous studies have suggested that most of the sites are edited at given that RNA editing correlates spatiotemporally with higher cog- low levels in all tissues or cell types (<20%)26,28,37. However, the lowly nitive functions, it would be informative to introduce specific RNA edited sites may be highly edited under different conditions. It appears editing sites into non-human primates and determine whether they that RNA editing is more frequent in brain tissues than in non-brain are sufficient to alter cognitive function. tissues, with more RNA editing sites and higher editing levels21,28. An RNA editing atlas will provide an unbiased view of RNA editing sites Interpretation of the functions of RNA editing sites at the transcriptome level in brain and non-brain tissues. In addition, RNA editing in the mRNA can yield multiple protein isoforms. The one can isolate different regions of the brain, which may be useful in vast majority of the editing events occur in non-coding regions. These linking RNA editing levels to specialized tasks governed by different events could be involved in many aspects of RNA metabolism, such as brain regions.

1520 VOLUME 16 | NUMBER 11 | NOVEMBER 2013 nature neuroscience r e v i e w

At the individual cell type or single cell level, an RNA editing that a number of cofactors are involved in fine-tuning RNA editing atlas will further reveal the dynamics of RNA editing that may not levels, giving rise to its wide spectrum of dynamics. Recently, primary be fully appreciated at the tissue or organ levels. For example, there screens using a yeast-based RNA editing assay led to the identifica- are hundreds of neuronal cell types with different functions in the tion of a handful of enhancers and repressors involved in ADAR RNA human brain. In animal models, RNA from particular cell types can editing48,49. However, such screenings using yeast are suboptimal as be isolated using a variety of techniques developed in mouse38, fly and a result of the lack of endogenous ADARs in yeast. Using metazoan worm39. Furthermore, single neurons can be isolated by laser capture systems in which ADARs are conserved, future studies may yield or microfluidic devices and analyzed by transcriptome sequencing. additional factors underlying the trans regulation of RNA editing. RNA editing might differ radically between two cells that are similar morphologically and adjacent anatomically, or even vary subcellularly. Conclusions The function of such variation may be more evident in the context Rapid technological developments have put us in an unprecedented of the cell types involved and/or their spatial relations. Single-cell position to study RNA editing. The DNA and RNA sequencing tech- RNA sequencing, fluorescent in situ sequencing of mRNA molecules40 nologies make it easy to identify and measure RNA editing in dif- and in vivo visualization of ADAR activity41 could help to illuminate ferent physiological and pathological scenarios. This will generate a such issues. These technologies may be in their infancy, but the low large number of hypotheses that correlate RNA editing with biological signal-to-noise ratio may be overcome by sampling a large number functions. Functional studies are made possible by perturbing the sta- of cells. As these technologies become more mature and are adopted tus of RNA editing using genome engineering and synthetic biology. by more research groups, a higher resolution atlas of RNA editing Although it is well established that A-to-I RNA editing is catalyzed will be achieved. by ADARs via binding to double-stranded RNA substrates, there is still much work to be done to define the code underlying the rules of Understanding cis and trans regulation of RNA editing cis and trans regulation. New approaches for systematically studying RNA editing is a dynamically regulated process, but its regulation is protein and RNA interactions will facilitate this endeavor. poorly understood. What determines which sites are edited and at The pursuit of RNA editing studies will be synergized by the recent what level? At the cis regulation level, the flanking sequences affect the initiative of the Brain Activity Map50. The new tools developed by selectivity of editing sites and the 5′ and 3′ flanking sequences appear the initiative will undoubtedly enable deeper understanding of RNA to be most influential42. However, these studies are often based on editing. Given the critical roles of RNA editing in brain activity, it is in vitro assays of a very limited number of substrates. In addition to also important to note that understanding the function and regula- the flanking sequence motifs, the local RNA structure is important, tion of RNA editing at single-nucleotide resolution may require the as the double-stranded RNA structure is required for ADAR bind- development of additional tools. ing. In addition to the stem-loop structure formed in the vicinity With the exciting development of technologies that are currently or of an editing site, a genomically distant RNA sequence or structure will soon be available, we expect that the prime time for studying RNA can also affect ADARs editing preference43. Looking forward, new editing is yet to come. The seemingly subtle differences introduced in in silico, in vitro and in vivo approaches are needed to determine the the transcriptome, which have been largely overlooked for decades, RNA and ADAR protein structures to characterize ADAR selectivity will continue to surprise us and lead us to fully appreciate it as a fine- or preference. tuning mechanism, particularly in the complex nervous system. At the trans regulation level, ADARs have an important role. © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature However, ADAR RNA and protein levels do not fully account for Acknowledgments We thank members of the Li laboratory for discussions and critical reading of 35 the spatiotemporal changes of RNA editing , suggesting that other the manuscript, and particularly R. Zhang for making Figure 2. This work was are involved in modulating ADAR activity. It is poorly under- funded by US National Institutes of Health (GM102484) and the Ellison Medical npg stood which other proteins beyond ADARs are involved in this regu- Foundation (to J.B.L.). lation. The transcription factor cAMP response element–binding COMPETING FINANCIAL INTERESTS protein, or CREB, can induce ADAR2 expression in hippocampal The authors declare no competing financial interests. CA1 neurons in rat brain44. Recently, two proteins, Pin1 and WWP2, were found to coordinately regulate ADAR2 post-translationally45. Reprints and permissions information is available online at http://www.nature.com/ Pin1 positively regulates ADAR2 activity by binding to ADAR2 in reprints/index.html. a phosphorylation-dependent manner. WWP2 negatively regulates ADAR2 by a protein interaction that results in ubiquitination and 1. Taft, R.J., Pheasant, M. & Mattick, J.S. The relationship between non-protein-coding subsequent degradation of ADAR2. In addition, two regulators have DNA and eukaryotic complexity. Bioessays 29, 288–299 (2007). 2. Nishikura, K. Functions and regulation of RNA editing by ADAR deaminases. been recently revealed using Drosophila. One is the Fragile X protein, Annu. Rev. Biochem. 79, 321–349 (2010). FMRP, which biochemically and genetically interacts with Drosophila 3. Palladino, M.J., Keegan, L.P., O’Connell, M.A. & Reenan, R.A. A-to-I pre-mRNA ADAR and subsequently modulates RNA editing levels46. The other editing in Drosophila is primarily involved in adult nervous system function and integrity. Cell 102, 437–449 (2000). is Period, which is involved in circadian rhythm. The editing levels 4. Tonkin, L.A. et al. RNA editing by ADARs is important for normal behavior in of a number of editing sites are substantially changed in the Period Caenorhabditis elegans. EMBO J. 21, 6025–6035 (2002). 47 5. Chen, C.X. et al. A third member of the RNA-specific adenosine deaminase gene loss-of-function mutant , although the relationship between ADAR family, ADAR3, contains both single- and double-stranded RNA binding domains. and Period awaits further study. RNA 6, 755–767 (2000). However, barring these few examples, the number and identity 6. Wang, Q., Khillan, J., Gadue, P. & Nishikura, K. Requirement of the RNA editing deaminase ADAR1 gene for embryonic erythropoiesis. Science 290, 1765–1768 of the other proteins involved in trans regulation remain mostly (2000). unknown. Thus, systematic approaches need to be developed to carry 7. Hartner, J.C. et al. Liver disintegration in the mouse embryo caused by out unbiased searches of additional regulators. The fact that some sites deficiency in the RNA-editing ADAR1. J. Biol. Chem. 279, 4894–4902 (2004). are specifically edited in brain or other tissues hints at the existence of 8. Higuchi, M. et al. Point mutation in an AMPA receptor gene rescues lethality in tissue-specific regulators that need to be identified. It is conceivable mice deficient in the RNA-editing enzyme ADAR2. Nature 406, 78–81 (2000).

nature neuroscience VOLUME 16 | NUMBER 11 | NOVEMBER 2013 1521 R e v i e w

9. Maas, S., Kawahara, Y., Tamburro, K.M. & Nishikura, K. A-to-I RNA editing and 31. Brusa, R. et al. Early-onset epilepsy and postnatal lethality associated with an human disease. RNA Biol. 3, 1–9 (2006). editing-deficient GluR-B allele in mice. Science 270, 1677–1680 (1995). 10. Burns, C.M. et al. Regulation of serotonin-2C receptor G-protein coupling by RNA 32. Olaghere da Silva, U.B. et al. Impact of RNA editing on functions of the serotonin editing. Nature 387, 303–308 (1997). 2C receptor in vivo. Front. Neurosci. 4, 26 (2010). 11. Kawahara, Y. et al. Dysregulated editing of serotonin 2C receptor mRNAs results in 33. Gaj, T., Gersbach, C.A. & Barbas, C.F. III ZFN, TALEN, and CRISPR/Cas-based energy dissipation and loss of fat mass. J. Neurosci. 28, 12834–12844 (2008). methods for genome engineering. Trends Biotechnol. 31, 397–405 (2013). 12. Silberberg, G., Lundin, D., Navon, R. & Ohman, M. Deregulation of the A-to-I 34. Mizrahi, R.A., Schirle, N.T. & Beal, P.A. Potent and selective inhibition of RNA editing mechanism in psychiatric disorders. Hum. Mol. Genet. 21, 311–321 A-to-I RNA editing with 2′-O-methyl/locked nucleic acid–containing antisense (2012). oligoribonucleotides. ACS Chem. Biol. 8, 832–839 (2013). 13. Eran, A. et al. Comparative RNA editing in autistic and neurotypical cerebella. 35. Wahlstedt, H., Daniel, C., Enstero, M. & Ohman, M. Large-scale mRNA sequencing Mol. Psychiatry 18, 1041–1048 (2012). determines global regulation of RNA editing during brain development. Genome 14. Chen, L. et al. Recoding RNA editing of AZIN1 predisposes to hepatocellular Res. 19, 978–986 (2009). carcinoma. Nat. Med. 19, 209–216 (2013). 36. Sanjana, N.E., Levanon, E.Y., Hueske, E.A., Ambrose, J.M. & Li, J.B. Activity- 15. Sommer, B., Kohler, M., Sprengel, R. & Seeburg, P.H. RNA editing in brain controls dependent A-to-I RNA editing in rat cortical neurons. Genetics 192, 281–287 a determinant of ion flow in glutamate-gated channels. Cell 67, 11–19 (1991). (2012). 16. Hoopengardner, B., Bhalla, T., Staber, C. & Reenan, R. Nervous system targets 37. Peng, Z. et al. Comprehensive analysis of RNA-Seq data reveals extensive RNA of RNA editing identified by comparative genomics. Science 301, 832–836 editing in a human transcriptome. Nat. Biotechnol. 30, 253–260 (2012). (2003). 38. Heiman, M. et al. A translational profiling approach for the molecular characterization 17. Levanon, E.Y. et al. Evolutionarily conserved human targets of adenosine to inosine of CNS cell types. Cell 135, 738–748 (2008). RNA editing. Nucleic Acids Res. 33, 1162–1168 (2005). 39. Steiner, F.A., Talbert, P.B., Kasinathan, S., Deal, R.B. & Henikoff, S. Cell type– 18. Levanon, E.Y. et al. Systematic identification of abundant A-to-I editing sites in the specific nuclei purification from whole animals for genome-wide expression and human transcriptome. Nat. Biotechnol. 22, 1001–1005 (2004). chromatin profiling. Genome Res. 22, 766–777 (2012). 19. Kim, D.D. et al. Widespread RNA editing of embedded alu elements in the human 40. Ke, R. et al. In situ sequencing for RNA analysis in preserved tissue and cells. transcriptome. Genome Res. 14, 1719–1725 (2004). Nat. Methods 10, 857–860 (2013). 20. Athanasiadis, A., Rich, A. & Maas, S. Widespread A-to-I RNA editing of Alu- 41. Jepson, J.E., Savva, Y.A., Jay, K.A. & Reenan, R.A. Visualizing adenosine-to-inosine containing mRNAs in the human transcriptome. PLoS Biol. 2, e391 (2004). RNA editing in the Drosophila nervous system. Nat. Methods 9, 189–194 (2012). 21. Li, J.B. et al. Genome-wide identification of human RNA editing sites by parallel 42. Eggington, J.M., Greene, T. & Bass, B.L. Predicting sites of ADAR editing in double- DNA capturing and sequencing. Science 324, 1210–1213 (2009). stranded RNA. Nat. Commun. 2, 319 (2011). 22. Li, M. et al. Widespread RNA and DNA sequence differences in the human 43. Daniel, C., Veno, M.T., Ekdahl, Y., Kjems, J. & Ohman, M. A distant cis transcriptome. Science 333, 53–58 (2011). acting intronic element induces site-selective RNA editing. Nucleic Acids Res. 40, 23. Kleinman, C.L. & Majewski, J. Comment on “Widespread RNA and DNA sequence 9876–9886 (2012). differences in the human transcriptome”. Science 335, 1302; author reply 1302 44. Peng, P.L. et al. ADAR2-dependent RNA editing of AMPA receptor subunit GluR2 (2012). determines vulnerability of neurons in forebrain ischemia. Neuron 49, 719–733 24. Pickrell, J.K., Gilad, Y. & Pritchard, J.K. Comment on “Widespread RNA and DNA (2006). sequence differences in the human transcriptome”. Science 335, 1302; author 45. Marcucci, R. et al. Pin1 and WWP2 regulate GluR2 Q/R site RNA editing by ADAR2 reply 1302 (2012). with opposing effects. EMBO J. 30, 4211–4222 (2011). 25. Lin, W., Piskol, R., Tan, M.H. & Li, J.B. Comment on “Widespread RNA and DNA 46. Bhogal, B. et al. Modulation of dADAR-dependent RNA editing by the Drosophila sequence differences in the human transcriptome”. Science 335, 1302; author fragile X mental retardation protein. Nat. Neurosci. 14, 1517–1524 (2011). reply 1302 (2012). 47. Hughes, M.E., Grant, G.R., Paquin, C., Qian, J. & Nitabach, M.N. Deep sequencing 26. Ramaswami, G. et al. Accurate identification of human Alu and non-Alu RNA editing the circadian and diurnal transcriptome of Drosophila brain. Genome Res. 22, sites. Nat. Methods 9, 579–581 (2012). 1266–1281 (2012). 27. Piskol, R., Peng, Z., Wang, J. & Li, J.B. Lack of evidence for existence of 48. Garncarz, W., Tariq, A., Handl, C., Pusch, O. & Jantsch, M.F. A high-throughput noncanonical RNA editing. Nat. Biotechnol. 31, 19–20 (2013). screen to identify enhancers of ADAR-mediated RNA-editing. RNA Biol. 10, 28. Ramaswami, G. et al. Identifying RNA editing sites using RNA sequencing data 192–204 (2013). alone. Nat. Methods 10, 128–132 (2013). 49. Tariq, A. et al. RNA-interacting proteins act as site-specific repressors of ADAR2- 29. Paz-Yaacov, N. et al. Adenosine-to-inosine RNA editing shapes transcriptome mediated RNA editing and fluctuate upon neuronal stimulation. Nucleic Acids Res. diversity in primates. Proc. Natl. Acad. Sci. USA 107, 12174–12179 (2010). 41, 2581–2593 (2013). 30. Lomeli, H. et al. Control of kinetic properties of AMPA receptor channels by nuclear 50. Alivisatos, A.P. et al. Neuroscience. The brain activity map. Science 339, RNA editing. Science 266, 1709–1713 (1994). 1284–1285 (2013). © 2013 Nature America, Inc. All rights reserved. America, Inc. © 2013 Nature npg

1522 VOLUME 16 | NUMBER 11 | NOVEMBER 2013 nature neuroscience